With　the　growth　of　e-commerce　systems,　the　recommendation　technology　has　made　great　success,　but　there　are　still　a　number　of　challenges,　including　the　problem　of　low　quality　and　data　sparseness.　In　order　to　improve　the　quality　of　the　recommendation,　we　put　the　timing　data　and　the　user＇s　register　information　into　the　traditional　collaborative　filtering　recommendation　algorithm　separately.　The　two　improved　algorithms　are　proposed　and　they　are　the　time　context　sensitive　algorithm　and　the　user　characteristic　information　sensitive　algorithm.　The　experimental　results　on　the　MovieLens　data　set　and　the　t-test　results　show　that　these　two　improved　algorithms　enhance　the　recommending　system　performance　significantly.　The　MAE　value　can　reach　0.7649　and　0.7603　separately.　©,　2015,　Binary　Information　Press.　All　right　reserved.